Stream

Sean Leather

Discussion  - 
 
A large collection of somewhat-organized snippets on a variety of Haskell-related topics.
1
Add a comment...
 
A useful Haskell cheatsheet.
 
This covers.. about everything. Nice cheatsheet.
http://dev.stephendiehl.com/hask/
4
Add a comment...

Andres Löh

Announcement  - 
 
 
Last chance! Next week are my #Haskell courses in London. Register now :)

2 days intro, 2 days advanced, and 1 day type magic: https://skillsmatter.com/explore?content=courses&q=haskell

Also, don't forget that there's a free "In the Brain" talk on Overloaded Record Fields in Haskell by my colleague Adam Gundry on April 28: https://skillsmatter.com/meetups/6345-overloaded-record-fields-for-haskell
1
Add a comment...

Oleg Durandin

Help and Advice  - 
 
Hi all!
Not long ago, I encouraged with the next chellange: I want to call from C/C++ Haskell function. I've already read a few tutorials, that related with this topic, but IO monad calls are not covered in these tutorials. In particular, I'd like to call function, that use "interact" function (interact :: (String -> String) -> IO () )

I do not understand how to use functions in this case?
How can I declarate function in C/C++ wrapper?
How control of standard input/output will be transfered between C/C++ and Haskell (in C/C++ code)?
1
Add a comment...

Andrew Cowie

Discussion  - 
 
I'm sure I'm missing something obvious, but why did Cabal's sandboxes go into .cabal-sandbox/ (hidden) and cabal.sandbox.config (not)? And meanwhile there's now a cabal.config too? Seems like at least the build debris could have ended up under the hideous dist/ tree, seeing as how we all have to see it, but surely all that fragmentation is a bit much?
8
Darren Grant's profile photo
 
Hadn't thought of it, but very good point. I wonder what motivated the current locations.

Maybe +Johan Tibell knows? 
Add a comment...

Eugene Crosser

Help and Advice  - 
 
I am trying to statically link a program that is dependent (indirectly) on a library written in C++. Linking step fails not finding dependencies that presumably should be present in libstdc++.

I'd much appreciate any Help and Advice how to make this work.

Thanks!
1
Eugene Crosser's profile photoArjun Comar's profile photo
3 comments
 
Thanks +Arjun Comar, "--pgml g++" is exactly what I was looking for! It does not help in my case, because (1) I need networking and getaddrinfo() does not work with statically linked programs (on different systems), and (2) static libicuuc complains about unresolved reference icudt52_dat.

But I managed to get rid of the dependency on icu: this dependency came from http-conduit / http-client via indirect dependency on publicsuffixlist (seriously?!). This latter is built with 'create' flag in the system package. When I rebuilt it locally without 'create', it no longer wanted icu libraries.

So in the end my program, dynamically linked but without dependency on icu, becomes "portable enough" as an executable.
Add a comment...
 
This is a very concise explanation I wrote up for how the `Functor`, `Applicative`, and `Monad` instances work for `ContT`.
I remember the first time I read the Monad instance for ContT I was so confused. I couldn't fathom how it worked because it was hard to discern the pattern. However, I later discovered that renaming things makes the pattern m...
12
Add a comment...

Bas van Dijk

Announcement  - 
 
ZuriHac 2014 is almost at capacity. Please register soon if you want to join.
2
Bas Dirks's profile photo
 
Does also being a Bas D. grant me free access (and a plane ticket)? :S
Add a comment...
9
3
Sergei Trofimovich's profile photoDarren Grant's profile photo
Add a comment...

Denis Shevchenko

Announcement  - 
 
Приветствую!

Рад сообщить, что первый том "О Haskell по-человески" теперь живёт онлайн: http://ohaskell.ru/fst.

Кстати, работа над вторым томом уже началась...
 ·  Translate
8
4
Semen Martynov's profile photoZalim Bashorov's profile photoFedor Isakov's profile photoDenis Shevchenko's profile photo
5 comments
 
Хм... Вы уже второй читатель, который говорит мне это... Думаю, вы правы, пример нужно чуток упростить.
 ·  Translate
Add a comment...

Chris Dornan

Announcement  - 
 
 
Fun and Profit with Stongly-Typed Data Schemas

I will be giving a talk in London on Wednesday evening at Skills Matter on api-tools, a cool package we at Iris Connect have been developing with Well Typed to help us with the new web platform. See the Skills Matter web site for details: https://skillsmatter.com/meetups/6201-fun-and-profit-with-stongly-typed-data-schemas
8
2
Ryan Newton's profile photoJim Stuttard's profile photoMark Mazumder's profile photo
 
I guess you know the tests currently fail (duplicate "generate" from QuickCheck and Tasty and something in Migration.hs)?
Add a comment...

Johan Tibell
moderator

Hackage  - 
 
Public Service Announcement

If you .cabal file has a flag that's only used for testing, make sure it has "manual: True", or cabal install will a) have a tougher job and b) might actually turn this flag on when installing the package.

Wrong:
flag developer
  description: operate in developer mode
  default: False

Right:
flag developer
  description: operate in developer mode
  default: False
  manual: True
1
Add a comment...

Sean Leather

Discussion  - 
 
A monospaced font with ligatures for Haskell.
7
1
Sean Leather's profile photoMattias Bengtsson's profile photoDarren Grant's profile photo
8 comments
 
+Richard Silverman What editor did you use?
Add a comment...

Mark Lentczner
moderator

Announcement  - 
13
2
Semen Martynov's profile photoSteven Grady's profile photoYun-Yan Chi's profile photoMark Lentczner's profile photo
2 comments
 
Yeah yeah... I first worried that it should be >>, or perhaps >=> (where the actions have type Human -> IO Human). Sure, those'd be more type correct... but >>= is just so iconic of Haskell, it had to be >>=.

But for those of you compiling my txts in your head, appreciate that Friday is capitalized, as it should be in a pattern match (before the ->), whereas saturday and sunday are not, as it would be pretty uncommon for constructors to be Alternatives.
Add a comment...

Johan Tibell
moderator

Announcement  - 
 
On behalf of all the cabal contributors, I'm proud to announce cabal 1.20. With 404 commits since the 1.18 release, this is a feature packed release. Here's a short summary of the perhaps most notable changes:

 * Dependency freezing
 * Parallel cabal build
 * Flag to temporary ignore upper bounds
 * Unnecessary re-linking avoidance
 * Streaming cabal test output
 * New cabal exec command
 * Haddock configuration options

To install:

cabal update
cabal install Cabal-1.20.0.0 cabal-install-1.20.0.0

I did a longer write-up about the new features on my blog: http://blog.johantibell.com/2014/04/announcing-cabal-120.html
42
3
Sergei Trofimovich's profile photoRyan Newton's profile photoDarren Grant's profile photoMichael Pavlovsky's profile photo
 
great news
Add a comment...

Bryan O'Sullivan
moderator

Discussion  - 
 
The optparse-applicative package is really cool, but made vastly more difficult to use by fragmented, incomplete documentation.

Riddle me this: if you're not the author of the package, can you find out, in less than 5 minutes, how to specify a default value for an option if none is supplied on the command line? I've provided a link to the documentation to, uh, help.
4
Alp Mestanogullari's profile photoAndrew Cowie's profile photo
10 comments
 
Note that we have hflags (a port of gflags) too: http://hackage.haskell.org/package/hflags
Add a comment...

Gergő Érdi

Announcement  - 
 
+Andy Gill : I have uploaded a consistent set of maintenance releases of Kansas Lava. They even support GHC 7.8.All you need to do is upload them to Hackage. What more can I do to get these on Hackage ASAP?
1
Add a comment...

Bryan O'Sullivan
moderator

Discussion  - 
 
My kingdom to whoever can cause Haddock to generate error messages better than the equivalent of "something went wrong lol kthxbai!"
9
Evan Laforge's profile photo
 
Doesn't the new version eliminate error messages entirely by making everything parse?  I suppose that means now you don't even get "something went wrong".
Add a comment...
 
Calling Python from Haskell 
10
1
John Millikin's profile photoRyan Newton's profile photoSebastian Wiesner's profile photo
20 comments
 
+John Millikin I can't help but find that balance a little odd, since in the case of cpython the underlying library deliberately chose to use a liberal license.

But it's your choice, and it's not up to me to pass judgement on it.

Thank you for your work, and for this discussion.
Add a comment...